In December 1995, the Hubble Space Telescope was trained on a small, undistinguished patch of the sky at high galactic latitude, and was left to expose nearly continuously for ten days. The resulting data set, a remarkable image known as the Hubble Deep Field (HDF), was distributed in both raw and reduced form to the wider astronomical community on 15 January 1996 to serve as a public dataset for observational cosmology. Within a single WFPC2 field of view, 2.5' on a side, the HDF images reveal roughly 3500 faint galaxies (and a few faint stars) down to approximately V = 30--the deepest optical images of the sky ever obtained. The 0.1" resolution of HST provides kpc-scale morphological detail for objects at all redshifts, permitting the sizes and structures of the faint galaxy population to be characterized.
The HST images are not only deep, but are in glorious technicolor as well, having been obtained through four WFPC2 passbands spanning the wavelength range 3000-8000 Å. For galaxies at z > 1, however, which should be numerous in the HDF, the HST images probe the rest-frame ultraviolet spectral region, where light from hot young stars dominates the emission. The optical wavelengths where we are most familiar with nearby galaxies are redshifted out into the near infrared. For this reason, deep IR imaging offers a valuable complement to the optical HST data, providing photometric information necessary to measure the light from cooler, older stellar populations at high redshift (see figure on the next page).
For this reason, 10 nights with the Mayall 4-m telescope were devoted to infrared imaging of the Hubble Deep Field. Principle investigator Mark Dickinson and a rotating cast of observers (Matt Bershady, Richard Elston, Adam Stanford, and Peter Eisenhardt) took turns making coffee and dithering the IRIM camera about the HDF. Good weather (largely photometric as well) held throughout the entire run.
IRIM on the 4-m offers an excellent match to the WFPC2 images, covering almost exactly the same field of view with its 256 x 256 NICMOS-3 array. This wide-field capability made the KPNO 4-m very well suited to the task of surveying the Deep Field. While no ground-based IR observations with present telescopes and instrumentation can go deep enough to detect all the galaxies found in the HST image, these KPNO data (probably the deepest images ever obtained with IRIM) provide valuable three band infrared photometry for hundreds of galaxies across the entire HDF.
Altogether, 22.9 hours of exposure were obtained on the HDF in the Ks-band, along with 11.3 and 11.0 hours at H and J, respectively. These exposure times reach similar depths, in the sense that most objects are detected with comparable S/N in all three passbands. Galaxies are visible to roughly K = 23, and the seeing in the grand-sum combined images is roughly 1" FWHM.
Following the lead set by STScI with the original HDF data, the IRIM images will be made available to the wider community after they have been reduced, calibrated and cataloged. Initially, the data will be distributed via anonymous ftp from NOAO and/or STScI. Eventually, it is hoped that they will be lodged in the HST Data Archive or another permanent distribution site. For further information, please see the HDF/IRIM WWW site accessible from the STScI HDF "follow-up clearinghouse" page, at http://www.stsci.edu/ftp/observer/hdf/clearinghouse/clearinghouse.html. Further information about this data and its distribution will be reported in the next NOAO Newsletter as well.
Mark Dickinson